Uncovering diagnostic biases

(This article was first printed in the June 2007 issue of the Harvard Mental Health Letter. For more information or to order, please go to http://www.health.harvard.edu/mental.)

Mental health professionals presented with a realistic, brief, fictional case history describing a troubled young man provided diagnoses that varied with the age and theoretical orientation of the clinician, the race of the client or patient, and the social context in which the case history was presented.

Nearly 1,400 psychiatrists, psychologists, and social workers, a representative professional sample, participated in the study. Researchers mailed each of them one of three case vignettes describing a boy with typical symptoms of conduct disorder: persistent behavior violating social norms and the rights of others, including deceit, theft, aggression, and rule violations like truancy and running away from home. Conduct disorder was chosen because it is the most common reason for referring adolescents for treatment.

In some case histories the young man was described as white, in others as black or Hispanic. The vignettes took one of three forms, depending on whether the symptoms were described as an internal problem or a reaction to a destructive environment. In a basic, neutral version, the researchers supplied only a description of the symptoms and information on the age, sex, and ethnic background of the patient. An “internal dysfunction” version included an added paragraph stating that the behavior appeared irrational, was indiscriminate, persisted when the environment changed, and suggested a lack of empathy or concern for the consequences. An “environmental reaction” version included a different paragraph suggesting that the young man lived in a dangerous neighborhood pervaded with gang violence and that his behavior changed when he left that neighborhood.

Each mental health professional was asked whether he or she would agree that a young man like this one had a mental illness or psychiatric disorder. Not surprisingly, the type of case history was the most important influence on this choice. Of those who read the neutral version, about 70% agreed that the young man had a conduct disorder. Of those reading the “internal dysfunction” version, nearly 95% agreed, and of those reading the “environmental reaction” version, only 35%.

The second most important influence was professional identity. Psychologists were twice as likely as social workers to diagnose a mental disorder, and psychiatrists nearly five times as likely.

Next came the race of the patient or client. Study participants were only 60% as likely to diagnose a mental disorder if the boy was labeled black or Hispanic. Then came age — older clinicians found mental disorders less often. Finally, clinicians who described their theoretical orientation as behavioral or eclectic were only about 65% as likely to diagnose a psychiatric disorder as were those who had a psychodynamic or psychoanalytic orientation.

Perhaps surprisingly, the race and gender of the clinician (about 50% were women and about 10% minority) had no effect on these judgments, even in combination with the race of the fictional client or patient.

The authors considered the reasons for these differences. Some, although not all, studies suggest that clinicians tend to find less pathology in blacks than in whites with similar symptoms. Black adolescents, in particular, are more often judged to be delinquent and assigned to juvenile justice rather than treatment. There is some evidence that racial effects on diagnosis can be eliminated by training in the use of the American Psychiatric Association’s diagnostic manual.

As for professional bias, psychiatrists are more familiar with the diagnostic manual and use it more often. Social workers in particular tend to focus on social context rather than internal states. Older clinicians may have diagnosed mental disorders less often because the manual, which relies almost entirely on symptoms alone for diagnosis, came into widespread use after their training was finished. The differences between psychodynamic and eclectic or behavioral therapists, according to the authors, probably reflect both self-selection and training.

	Click to enlarge		Harvard Mental Health Letter The Harvard Mental Health Letter is a unique resource that covers a wide range of mental health issues and concerns. It presents the latest thinking, treatment options, therapies and debate of interest to both mental health care professionals and the concerned public. Read more 12 monthly issues (Print+Electronic) $59.00 12 monthly issues (Electronic Only) $55.00